A to Z (NLP) Machine Learning Model building and Deployment.

Explore sentiment analysis with nlp: classify tweets as negative, neutral, or positive using a python notebook, building and evaluating a model to improve accuracy.

Develop a Flask based API in Python to deploy an MLP model and serve sentiment estimates, using regex text cleaning, punctuation counting, and a logistic regression classifier.

Learn how to preprocess text for NLP by removing punctuation and special characters, applying tokenization and stemming, and computing body length and punctuation features to boost model performance.

Learn hyperparameter tuning and model selection for nlp using tf-idf features, grid search cv, and logistic regression to improve accuracy, with plans to deploy via flask and docker.

Explore how a Flask app uses templates and static files to render a sentiment analysis interface, with a home page, a form, and a /predict route showing sentiment results.

Launch and test a Flask API by running Python app.py, expose it on localhost:4000, and validate sentiment predictions on sample tweets with a dockerized deployment workflow.

Create a dockerfile to package a Flask API for the MLP model, using an Alpine base image, install dependencies, copy the app to /app, and expose port 4000.

Copy your local files to a remote environment using winSCP by logging in with hostname, port 22, and credentials, then drag and drop files to the remote directory.

Create a Jenkinsfile to define a pipeline with stages for cloning the repo, building and running a docker image, and executing tests, including permissions troubleshooting.

Configure a ci-cd pipeline by setting up a GitLab webhook in Jenkins, enable push and merge request triggers, and test the docker-based build pipeline end-to-end.

Celebrate completing the course and master the end-to-end NLP machine learning model building and deployment pipeline with hands-on tools.